Is grug-9b or gemma-4-12B-it smaller?

grug-9b is smaller at Q4_K_M: 5.24 GiB against 6.63 GiB.

Which uses less memory at long context?

grug-9b — its KV cache at 32K context is 2.5x smaller, because of how its attention layers cache.

Model comparison

grug-9b vs gemma-4-12B-it

At Q4_K_M, grug-9b is the smaller download — 5,629,109,056 bytes against 7,121,861,440. At long context the gap widens: grug-9b's KV cache at 32K is 2.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	grug-9b	gemma-4-12B-it
Parameters	9.4B	12.0B
Architecture	qwen35	gemma4
Layers	32	48
Native context	262,144	262,144
Mixture of experts	no	no
Quantizations published	28	59
Smallest quantization	3.51 GiB	3.92 GiB
Q4_K_M	5.24 GiB	6.63 GiB
Licence	mit	apache-2.0

KV cache by context

the term that decides long-context viability

Context	grug-9b	gemma-4-12B-it	Ratio
4,096	0.13 GiB	0.72 GiB	5.75×
8,192	0.25 GiB	0.97 GiB	3.87×
16,384	0.50 GiB	1.47 GiB	2.94×
32,768	1.00 GiB	2.47 GiB	2.47×
65,536	2.00 GiB	4.47 GiB	2.23×
131,072	4.00 GiB	8.47 GiB	2.12×